[OSM-talk] licence plan - Question about supplying own data

[SteveC]

On 4 Mar 2009, at 18:03, Frederik Ramm wrote:

> Hi,
>
> SteveC wrote:
>> One of the things I didn't mention in my long post an that most of  
>> you clearly don't understand is that a court takes *intent* in to  
>> account as guidence in any license dispute. So like case law you  
>> can spend all the time you like reading the letter of the license  
>> but if the intent was clearly elsewhere *it doesn't matter*.
>
> All the time you say that we mere humans should not talk legalese  
> but leave this to the lawyers.

No not really... I'm suggesting that you're intelligent and have great  
skills but those skills aren't intellectual property law, and its  
better that if a lawyer has a code question he asks you, and if you  
have a law question you ask them. Your time is way better spent that  
way.

> But if it were all about intent, then we could just write a document  
> in plain English and use that.

And there seems to be your misunderstanding of law. 'plain English'  
often just isn't, every word has special meanings which are open to  
interpretations you and I just don't know about.

> In fact I think that in many jurisdictions this would actually give  
> us the same level of "protection" than a longish "license" with the  
> added benefit that everybody can understand our intent, whereas -  
> proof is on odc-discuss where there are some people who read the  
> ODbL out of OSM context! - it seems that reading ODbL is not enough  
> to understand our intent.

Exactly. Reading the ODbL gives you one 30,000 ft viewpoint but for a  
full understanding you need to brake open your case of IP lawyers and  
ask them to look. As soon as you get to that level, its best to hand  
over the baton to them.

>>> That "feature" is something that was introduced without so much as a
>>> word from anyone between the April 2008 and the 0.9 drafts. If  
>>> this were
>>> intentional, then someone had to hang for trying to deceive the
>>> community.
>> You're doing it again - jumping to the conclusion that it was all  
>> Evil Jordan or Evil Steve.
>
> No. I am truly, honestly assuming that this is a blunder that nobody  
> noticed. The previous license draft said quite clearly that if you  
> publicly "use" a database, which included making a Produced Work (it  
> was called differently then but that doesn't matter) then you would  
> have to make the database available.
>
> The new version introduced the word "convey" instead of "use", which  
> is not a big deal, but then defined "convey" as "not applying to  
> making Produced Works".

My memory about this is that specific change is because of some recent  
case law which Clark picked up on. If you have a specific question  
(I'm not entirely sure what you're asking...) lets queue that up for  
legal? I think I know the issue and I thought it had been fixed.

> In doing so, the license went 180° - from forcing people to make  
> available a derived database on which they built Produced Works to  
> not forcing them to do so. All the time, proponents of the license  
> change (which, you might be surprised to see, include me), argued to  
> the hardcore share-alikers: "You don't get protection of Produced  
> Works but you get share-alike for interim databases which is much  
> more in line with what we want". This has been, and still is, a  
> fundamental point, an argument without I could not justify ODbL to  
> these people.
>
> It seems absolutely inconceivable that someone *knowingly* made such  
> a huge change and not even bothered to tell anyone. Even with my  
> mild degree of paranoia, it still seems absolutely inconceivable.
>
> That's why I write "if this were intentional, someone had to hang  
> for trying to deceive the community". If someone, anyone, really  
> thought "hey, I'll make this change and nobody will notice that this  
> will nicely drop database share-alike for improved data on which  
> Produced Works are built, so that I can, in the future, add my own  
> improvements, publish cool images, and never share my improvements";  
> if anyone really was that devious, that would be absolutely beyond  
> the pale and that person should not be allowed to say the words  
> "OpenStreetMap" or "community" ever again.
>
> But believe it or not, I don't think that there was such a secret  
> agenda on anyone's part. I think it simply has been overlooked. I  
> have just spent two weeks amid printed paper finalising the second  
> edition of our OSM book and I know perfectly well how easy it is to  
> change a meaning on page 3 by rewording something on page 1, and not  
> noticing it until the proofreader paints it red.
>
> The ODbL has been proofread, the problem has been painted red, and I  
> am sure it will be fixed. It *must* be fixed, because otherwise ODbL  
> is dead for OSM; you can claim that intent matters all you want, but  
> you will never be able to get a license past the Share-Alike  
> fraction that allows improvements to our data to be kept privy.

Okay... can you summarise this in to a question or set of questions  
that can be passed on with a paragraph of explanation that's as  
concise as possible? There seem to be multiple things within this and  
I want to make sure we ask the right things and not confuse it.

Best

Steve